Background Arrhythmogenic cardiomyopathy (ACM) is a leading cause of sudden cardiac death in young adults and athletes. It is mostly genetically determined, with causal variants predominantly affecting desmosomal genes. Approximately one-third of patients remain gene-elusive despite extensive testing. Gene-elusive patients often report significantly higher lifetime exercise exposure than those with desmosomal ACM, suggesting a disproportionate role of exercise in this subgroup, also termed ‘exercise-induced ACM’. Yet, the effect of exercise history on the clinical characteristics and prognosis of this phenotype remains scarce.Methods In this retrospective, single-centre cohort study, we included patients with definite or borderline ACM and negative genetic testing for ACM-associated genes. Exercise exposure was quantified with a standardised sports activity survey or from medical records. Competitive athletes were defined by an exercise exposure ≥24 metabolic equivalent of task-hour/week for ≥6 consecutive years and regular participation in competitive events. Life-threatening ventricular arrhythmias (LTVA) were defined as sustained ventricular tachycardia, ventricular fibrillation or appropriate implantable cardioverter-defibrillator therapies.Results 46 patients (26 athletes, 20 non-athletes) were followed for a median of 7.5 years. At diagnosis, most athletes (77%) had definite right-ventricular ACM, compared with only 40% of non-athletes (p=0.025). Compared with non-athletes, athletes had a larger end-diastolic right-ventricular volume, while other imaging findings did not differ significantly. Athletes showed a distinct electrical phenotype with a shorter QTc (424 ms vs 449 ms, p<0.001), more T-wave inversions (4.0 vs 2.40, p=0.047) and a higher hazard of LTVA, death or heart transplantation during follow-up (HR 2.67, 95% CI 1.11 to 6.44, p=0.029).Conclusion In gene-elusive ACM, a history of competitive athletic participation was associated with more extensive repolarisation abnormalities, larger right ventricular volumes and a higher hazard of adverse outcomes during follow-up. These associations require validation in prospective multicentre cohorts, particularly regarding the relationship between exercise exposure and arrhythmic risk.